JP2017140595A - Management support system, management support method and management support program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a management support system, a management support method and a management support program capable of supporting injection of a proper amount of a chemical in a water processing plant.SOLUTION: A management support system of the present embodiment includes a statistical prediction part, a water quality reaction model prediction part, and a management support part. The statistical prediction part statistically processes water quality information, chemical information and processed water quality information of processed water to acquire a statistical injection ratio, and predicts statistically processed water quality information based on water quality information of raw water that is a prediction target, a statistical injection ratio, and specified relation. The water quality reaction model prediction part utilizes a water quality reaction model modeling a reaction due to a chemical to acquire a model injection ratio, and based on the water quality information, the model injection ratio, and the water quality reaction model, model processed water quality information. The management support part generates an operation condition for controlling the injection ratio of the chemical based on the statistical injection ratio and the model injection ratio.SELECTED DRAWING: Figure 2

Description

  Embodiments described herein relate generally to a management support system, a management support method, and a management support program.

  The main purpose of the operation management of the water purification process at the water purification plant is to maintain the quality of the water after the water purification treatment so that it falls below the water quality standard for tap water. And each water purification plant generally has a higher target water quality (a lower water quality in terms of concentration) than the water quality standards, and then costs for cleaning and sludge disposal. As for energy and energy, efforts are being made to achieve energy and cost savings. In particular, in recent years, low-cost operation management technology has been demanded from the viewpoint of strengthening the operational infrastructure of water utilities.

  Generally, in a water treatment plant such as a water purification plant, there is a process for settling and removing suspended suspended matters contained in raw water. This is called a solid-liquid separation process. This solid-liquid separation process includes an agglomeration process in which a chemical called an aggregating agent is injected to form flocs and increase the sedimentation rate in order to increase the efficiency of removing suspended suspended matters from water. A floc is formed by agglomerating suspended suspension and flocculant. Flocks vary in size from what is called tens of micrometers, which cannot be visually confirmed, to several millimeters that grow until they can be observed with the naked eye, to what is called a giant flock of several centimeters. is there. A good agglomeration process forms a good floc. In order to form a good floc, it is important to consider not only the amount of the flocculant injected into the suspended suspension, but also water quality parameters such as pH, alkalinity and water temperature of the raw water. A good floc means a floc having a high density and a large particle size. If the density of floc is high and the particle size is large, the sedimentation property is improved and solid-liquid separation is promoted.

  Determination of the injection rate of chemicals such as flocculants used in water purification plants is often adjusted based on the experience and know-how of skilled workers, and it is difficult to inherit the technology. As a countermeasure, there is a technique of predicting and guiding a chemical injection rate by statistically processing past water quality data or actual data of chemical injection, or mechanically learning. In recent years, techniques for analyzing the relationship between chemical injection and water quality using big data analysis have also been proposed. However, chemical injection at a water purification plant may be excessively injected due to wasteful operation at that time, or due to circumstances of the water purification plant that wants to avoid troubles due to insufficient injection. There are many cases. For this reason, when machine learning or big data analysis is performed based on past data to predict the injection rate of the medicine, the predicted injection rate becomes a value including the above-described waste and excess. Injecting more than necessary chemicals becomes an impediment to reducing the cost of water treatment plant operation.

JP 2000-218263 A Japanese Patent Laid-Open No. 2007-061800

  The problem to be solved by the present invention is to provide a management support system, a management support method, and a management support program capable of providing support for injecting a more appropriate amount of chemicals in a water treatment plant.

  The management support system of the embodiment includes a statistical prediction unit, a water quality reaction model prediction unit, and a management support unit. The statistical prediction unit includes, as information acquired in the water treatment plant, water quality information relating to the quality of raw water, chemical information relating to injection of chemicals into the raw water, and processing after treatment with the chemicals injected according to the chemical information. Processed water quality information indicating the quality of the water is obtained and statistically processed to identify the relationship between the chemical information and the treated water quality information. Based on the identified relationship, the quality of the treated water is predetermined. The statistical injection rate, which is a lower injection rate in the range, is obtained, and statistical processing is performed based on the water quality information of the raw water to be predicted, the acquired statistical injection rate, and the specified relationship. Predict water quality information. The water quality reaction model prediction unit uses a water quality reaction model that models the reaction by the injected chemicals, and obtains a model injection rate that provides a lower injection rate while keeping the quality of the treated water within a predetermined range. The model treated water quality information is predicted based on the water quality information of the raw water to be predicted, the acquired model injection rate, and the water quality reaction model. The management support unit, based on the statistical injection rate used for prediction by the statistical prediction unit and the model injection rate used for prediction by the water quality reaction model prediction unit, injection of chemicals in the water treatment plant Generate operating conditions to control the rate.

The figure which shows the structural example of the management assistance system of the water purification plant and water purification plant in this embodiment. The figure which shows the detailed structural example of the management assistance system 20 in this embodiment. The figure which shows the outline | summary of the process in the mixing basin model part 201 of this embodiment. The figure which shows the outline | summary of the process in the flock formation pond model part 202 of this embodiment. The figure which shows the outline | summary of the process in the sedimentation basin model part 203 of this embodiment. The figure which shows the specific example of the model which expressed the water quality reaction more concretely in the water quality reaction model prediction part 23 of this embodiment. The flowchart which shows operation | movement of the management assistance system 20 in this embodiment.

  Hereinafter, a management support system, a management support method, and a management support program according to embodiments will be described with reference to the drawings.

First, an outline of a water purification plant and a water purification plant management support system in the present embodiment will be described.
FIG. 1 is a diagram illustrating a configuration example of a water purification plant and a water purification plant management support system according to the present embodiment. As shown in FIG. 1, the water purification plant 1 is a water storage facility for treating raw water, such as a landing well 3, an activated carbon contact basin 4, a mixing basin 5, a flock formation basin 6, a sedimentation basin 7, a sand filtration basin 8, and a clean water. A pond 9 is provided. The water purification plant 1 further includes flow meters 2a to 2d, a sampling pump 4a, a water temperature meter 10, a turbidity meter 11, a PH meter 12, a flow meter 13, a sampling pump 7a, and a sedimentation tank outlet turbidity. A total 14, an injection amount control unit 15, a flocculant injection facility 16, and a plant operation unit 26 are provided.

  The management support system 20 includes a data collection / storage unit 21, a statistical calculation unit 22, a water quality reaction model prediction unit 23, a parameter adjustment unit 24, and an operation management support unit 25. The data collection / storage unit 21 includes data measured by the flow meters 2a to 2d, the water temperature meter 10, the turbidity meter 11, the PH meter 12, the flow meter 13 and the sedimentation tank outlet turbidity meter 14 (hereinafter referred to as plant data). Collect and save. The management support system 20 has a function of transmitting information on the injection rate of chemicals such as a flocculant to the plant operation unit 26 based on the plant data collected in the data collection / storage unit 21, for example, a plurality of computers Consists of terminals.

  In the water treatment plant 1, the water intake process using the landing well 3, the activated carbon adsorption process using the activated carbon contact pond 4, the flocculant injection process using the mixing pond 5, the flock formation process using the flock formation pond 6, and the sedimentation tank 7 A plurality of water purification processes, such as a precipitation filtration process using a sand filter and a sand filtration process using a sand filtration pond 8, are performed.

  In the water intake process, for example, raw water is taken from a plurality of water intake sources A to D and flows into the landing well 3 through pipes. The raw water from each of the water intake sources A to D is measured by flow meters 2a to 2d provided in the respective pipes. Water intake flow rate data from the flow meters 2a to 2d is sent to the data collection / storage unit 21 as plant data.

  The raw water taken into the landing well 3 flows into the activated carbon contact pond 4 through the pipe. The raw | natural water which flowed in is processed by deodorization etc. by the activated carbon adsorption process in the activated carbon contact pond 4. FIG. In the activated carbon contact pond 4, the raw water sample is taken out by the sampling pump 4 a, and the water temperature, turbidity, and pH of the raw water are measured by the water temperature meter 10, the turbidity meter 11, and the PH meter 12. These measured values are sent to the data collection / storage unit 21 as plant data. Here, the case where activated carbon is introduced is a case where odor countermeasures such as mold odor and removal of colors such as chromaticity are required. Some water treatment plants have activated carbon facilities but do not always inject them.

  From the activated carbon contact basin 4, it flows into the mixing basin 5 through a pipe. Inflow flow rate data measured by a flow meter 13 provided in the pipe is sent to the data collection / storage unit 21 as plant data. In the mixing basin 5, a flocculant injection process in which the flocculant is injected from the flocculant injection equipment 16 is executed. In the flocculant injection facility 16, the injection rate of the flocculant is controlled by the operation management support unit 25 through the injection amount control unit 15 and the plant operation unit 26 as described later.

  The raw water into which the flocculant has been injected in the mixing basin 5 collides with the turbidity in the raw water and the flocculant, and is agglomerated by attracting charges to form micro flocs. Thereafter, it is sent to the flock formation pond 6. In the flock formation pond 6, a flock formation process is performed in which flocks are formed by agglomeration of micro flocs, residual flocculant, residual turbidity, and the like contained in the raw water. Furthermore, in the sedimentation basin 7, turbidity removal of raw | natural water is performed by sedimentation separation by a precipitation filtration process.

  At the outlet of the sedimentation basin 7, a sample of treated water is taken out by the sampling pump 7a, and the sedimentation turbidity at the outlet of the sedimentation basin 7 is measured by the sedimentation basin outlet turbidimeter 14. The measured value data of the sediment water turbidity at the outlet of the sedimentation tank 7 is accumulated in the data collection / storage unit 21 as plant data.

  The data collection / storage unit 21 stores, as time series data, water quality information, chemical information, and turbidity (precipitation water turbidity) at the outlet of the settling basin 7 as a processing result. The raw water quality information includes raw water turbidity, raw water pH, raw water temperature, and raw water alkalinity. As chemical | medical agent information, the information regarding injection | pouring rates [mg / L], such as a flocculant, a pH adjuster, and an alkalinity adjuster, is included. The chemical information includes information such as the type of flocculant and the basicity of aluminum in the case of an aluminum-based flocculant. The data collection / storage unit 21 stores, as structural information of the plant, a G value representing the stirring intensity, here, the G value of the mixing basin 5 and the G value of the flock formation pond 6. Since structural information is small when it is variable, it is often a constant value unless construction such as updating is performed.

  The data collection / storage unit 21 also stores raw water flow rate data collected from the flow meters 2 a to 2 d and the flow meter 13. The flow rate of raw water at the water purification plant may be constant for 24 hours or may vary depending on the demand for tap water. By using the flow rate of the raw water and the structural volume of the plant, the residence time of the raw water in each pond can be calculated. When this residence time is t, an index called a Gt value can be obtained by multiplying the G value described above. This Gt value is used as an index indicating how much strength the raw water has been subjected to stirring. The data collection / storing unit 21 also saves these indexes. In addition, the data collection / storage unit 21 stores information on the injection rate of the flocculant, the injection rate of the pH adjuster, and the injection rate of the alkalinity adjuster. In addition, a pH adjuster and an alkalinity adjuster are inject | poured into the landing well 3 which is a process before the mixing basin 5 in order to adjust pH and alkalinity of raw | natural water, for example. By this adjustment, the pH and alkalinity of the raw water are set to values suitable for floc formation.

  The data collection / storage unit 21 also stores water quality data of treated water as a result of performing each process. The data collection / storage unit 21 stores, for example, the turbidity of the precipitated water at the outlet of the sedimentation basin 7, the turbidity of the filtered water at the outlet of the filtration basin 8, and the water level rise rate (filter resistance rise) of the filtration basin 8. The data collection / storage unit 21 may be configured to store the amount of sludge when the amount of sludge generated in that period is calculated from the amount of sludge extracted in the settling tank 7 for a certain period.

  The statistical calculation unit 22 has a function of calculating treated water quality (statistically treated water quality information) including precipitated water turbidity by statistical processing using data stored in the data collection / storage unit 21. For example, the statistical calculation unit 22 classifies the data stored in the data collection / storage unit 21 according to the water quality pattern of the raw water. When the raw water quality information and the chemical injection rate are input, the statistical calculation unit 22 extracts past raw water quality having similar water quality patterns. The statistical calculation unit 22 obtains a relational expression between the chemical injection rate and the precipitated water turbidity based on the past results of the chemical injection rate and the precipitated water turbidity stored corresponding to the extracted raw water quality. The statistical calculation unit 22 predicts statistical treatment water quality information that is a treatment water quality (precipitation water turbidity) according to the input chemical injection rate based on the raw water quality information and this relational expression. The statistical calculation unit 22 compares the predicted statistical treatment water quality information with the actual treatment water quality stored in the data collection / storage unit 21, and based on the comparison result, the chemical injection rate and the sedimentation turbidity A learning process for correcting the degree relational expression is performed. The statistical calculation unit 22 generates a relational expression with higher prediction accuracy by performing such learning processing.

  The statistical calculation unit 22 performs the following processing for obtaining an optimum injection rate of the medicine using the above-described relational expression based on the water quality information of the raw water to be injected with the medicine. The statistical calculation unit 22 predicts a plurality of pieces of statistically treated water quality information based on the current water quality information of the raw water and the injection rate of the chemicals set to any plurality of types of values. The statistical calculation unit 22 outputs the lowest chemical injection rate as the statistical injection rate among the predicted plurality of statistically processed water quality information within the allowable range. This statistical injection rate is an optimal (for example, the lowest) chemical injection rate within a predetermined range of statistically processed water quality information obtained by the statistical calculation unit 22 through statistical processing. An example of a method for obtaining the lowest chemical injection rate is shown below. First, the statistical calculation unit 22 predicts statistically treated water quality information based on the injection rate of the medicine as an initial value. The initial value of the chemical injection rate is a value at which the predicted statistically treated water quality information is outside the allowable range. Thereafter, the chemical injection rate is increased by a predetermined value, and the statistical calculation unit 22 repeats the process of predicting statistically processed water quality information. When the statistical calculation unit 22 determines that the statistically processed water quality information predicted based on the current raw water quality information and the chemical injection rate set to an arbitrary value has reached the target level. The chemical injection rate is defined as the statistical injection rate. The statistical calculation unit 22 has a function of outputting operation conditions including a statistical injection rate to the operation management support unit 25. If the statistical calculation unit 22 predicts that the statistically processed water quality information does not reach the target level, for example, the statistically processed water quality information predicted again by increasing the injection rate of the input chemical by one step. Determine whether or not the target level is reached. The output operation condition is transmitted to the driving management support unit 25.

  Note that the statistical calculation unit 22 may use a technique such as principal component analysis or principal component regression analysis in the prediction of statistically processed water quality information. This method is a multivariate statistical process monitoring (MSPC: Multi-statistical analysis method) that uses the “multivariate statistical analysis method” that has been developed mainly in the petrochemical process field as a method for quickly grasping changes in the process state. It uses a method called Variate Statistical Process Control. In MSPC, a monitoring method using principal component analysis, principal component regression analysis, latent variable projection method / partial least square method, or the like is used. These methods are methods in which several pieces of statistical data are generated from many pieces of measurement data using correlation information between many pieces of process data, and changes in the process state are detected based on the generated pieces of statistical data. . In addition, the statistical calculation unit 22 may predict the statistically treated water quality information using another known method. For example, the statistically processed water quality information may be predicted using the methods described in Patent Documents 1 and 2 described above. It is good also as a structure which predicts.

  The water quality reaction model predicting unit 23 predicts model treated water quality information, which is the treated water quality of the virtually treated water based on a model obtained by more specifically formulating the water quality reaction in the water purification plant. The water quality reaction model prediction unit 23 defines a water quality reaction model of the flocculation process, and models the result of the treated water quality when selecting the injection rate of chemicals such as an arbitrary flocculant with respect to the water quality information of the raw water. Predicted as treated water quality information. The water quality reaction model prediction unit 23 receives the raw water quality information and the chemical injection rate, and treats the treated water quality including the precipitated water turbidity and the precipitated water aluminum concentration predicted based on the water quality information and the chemical injection rate. Model water quality information is output.

  The parameter adjustment unit 24 compares the water quality reaction model prediction unit 23 with the input values (water quality information and chemical injection rate) used when the statistical calculation unit 22 predicts the statistically processed water quality information. Instructs to predict the treated water quality information. The parameter adjustment unit 24 compares the comparative treatment water quality information predicted by the water quality reaction model prediction unit 23 according to the instruction and the statistical treatment water quality information predicted by the statistical calculation unit 22, and the two values are different. Judge whether or not. For example, the parameter adjustment unit 24 determines whether the difference value between the comparison process water quality information predicted by the water quality reaction model prediction unit 23 and the statistical process water quality information predicted by the statistical calculation unit 22 exceeds a predetermined threshold. It is determined whether or not there is a deviation depending on whether or not.

  When the parameter adjustment unit 24 determines that there is a divergence, the parameter used by the water quality reaction model prediction unit 23 is changed using the data of the data collection / storage unit 21, and the new parameter is changed to the water quality reaction model prediction unit 23. Set to. The parameter adjustment unit 24 adjusts until obtaining a parameter for which it is determined that the comparatively processed water quality information predicted by the water quality reaction model prediction unit 23 and the statistically processed water quality information predicted by the statistical calculation unit 22 are not deviated. repeat.

  If the parameter adjustment unit 24 determines that there is no deviation, the parameter adjustment unit 24 instructs the water quality reaction model prediction unit 23 to calculate the lowest chemical injection rate within a range in which the model treated water quality information is maintained at a predetermined level. To do. Based on the water quality reaction model, the water quality reaction model predicting unit 23 calculates a model injection rate that is the lowest chemical injection rate among the water quality information of the raw water and the model water quality information of the treated water is within the allowable range. Can be sought. This model injection rate is the optimal (for example, the lowest) chemical injection rate within the predetermined range of the treated water model treated water quality information obtained by the water quality reaction model prediction unit 23 based on the water quality reaction model. is there. The water quality reaction model prediction unit 23 outputs operation conditions including the calculated chemical model injection rate to the operation management support unit 25. The statistical calculation unit 22 and the parameter adjustment unit 24 periodically perform the above-described operation. Note that the statistical calculation unit 22 and the parameter adjustment unit 24 are not limited to a configuration that operates periodically, but monitors changes in the water quality information stored in the data collection / storage unit 21 and the water quality information changes greatly. In this case, the above-described operation may be performed.

  The operation management support unit 25 has a function of acquiring a target water quality level of the treated water quality. The operation management support unit 25 generates an optimal operation condition within a range in which the target water quality level is achieved based on the operation condition from the statistical calculation unit 22 and the operation condition from the water quality reaction model prediction unit 23. To the plant operation unit 26.

  The plant operation unit 26 automatically controls the injection rate of the flocculant according to the operation conditions from the operation management support unit 25. Specifically, the plant operation unit 26 outputs a control signal for controlling the injection rate of the flocculant to the injection amount control unit 15 according to the injection rate included in the operation condition from the operation management support unit 25. The injection amount control unit 15 controls the injection amount of the flocculant injected into the mixing basin 5 in the flocculant injection facility 16 according to the control signal received from the plant operation unit 26.

  The operation management support unit 25 and the plant operation unit 26 are configured to be communicably connected via a network. The operation management support unit 25 and the plant operation unit 26 are not limited to the configuration that enables communication via a network. The operation management support unit 25 may be configured to present operation conditions to an operator who operates the plant operation unit 26 (for example, displayed on a portable terminal held by the operator). Thereby, even if it is the structure which does not connect between the operation management support part 25 and the plant operation part 26 by a network etc., an operator performs operation according to the presented operation condition with respect to the plant operation part 26. be able to.

  With the above configuration, the management support system 20 can perform support for injecting a more appropriate amount when injecting the flocculant in the water purification plant 1. Since the statistical calculation unit 22 calculates the statistical injection rate based on the past data, the statistical calculation unit 22 shows data indicating how far the injection rate of the flocculant can be lowered while maintaining the statistically treated water quality information at a predetermined level. Is difficult. However, the water quality reaction model predicting unit 23 can obtain the model injection rate of the flocculant that is the lowest within the range in which the model process water quality information is set to a predetermined level based on the water quality reaction model. The management support system 20 can instruct the plant operation unit 26 of the water purification plant 1 to inject the flocculant injection rate in consideration of the model injection rate obtained by the water quality reaction model prediction unit 23.

Next, a detailed configuration of the management support system 20 in the present embodiment will be described.
FIG. 2 is a diagram illustrating a detailed configuration example of the management support system 20 in the present embodiment. As shown in FIG. 2, the water quality reaction model prediction unit 23 of the management support system 20 includes a mixing basin model unit 201, a flock formation pond model unit 202, a sedimentation basin model unit 203, a filtration pond model unit 204, and sludge. An information acquisition unit 205 and an operation condition determination unit 206 are provided. The parameter adjustment unit 24 includes a parameter history storage unit 240 that stores parameters previously set in the water quality reaction model prediction unit 23.

  FIG. 3 is a diagram showing an outline of processing in the mixing pond model unit 201 of the present embodiment. The mixing pond model unit 201 models and uses the water quality reaction of the coagulation process in the mixing pond 5. As shown in FIG. 3, the mixing pond model unit 201 calculates the charge neutralization rate and the surplus aluminum concentration based on the pH and alkalinity of the raw water, the injection rate of the flocculant, and the turbidity of the raw water, Output.

  In the mixing pond 5, when a flocculant is added to raw water, the flocculant is in a state where different forms are mixed. The flocculant often used in water purification plants is a polymer called an aluminum-based flocculant, commonly called PAC (polyaluminum chloride). When this PAC is added to raw water, it does not take a single structure, but a mixed state of three types of aluminum (hereinafter referred to as the first form of aluminum to the third form of aluminum). . The mixing ratio of the first form of aluminum to the third form of aluminum varies depending on the pH and alkalinity of the added raw water.

  The first form of aluminum is a relatively small form that does not contribute much to the agglomeration reaction. The aluminum in the second form is a form in which the PAC has undergone multivalent ionic polymerization, and this form of aluminum is said to be effective for charge neutralization. The aluminum of the third form is a form containing relatively large insoluble aluminum. This form of aluminum is considered to contribute greatly to the crosslinking action in the agglomeration reaction, and is a form that plays an important role in achieving a state in which flocs become large and easily settle. If this insoluble aluminum is insufficient, the charge neutralization progressed well, but the crosslinking action did not work well, so a large floc could not be made, and the sedimentation of the floc was poor, so the turbidity did not fall Will occur.

  The ratio of the first form of aluminum to the third form of aluminum when the aluminum-based flocculant as described above is poured into raw water is affected by the pH and alkalinity when added to water. The mixing pond model unit 201 calculates the ratio of the first form of aluminum to the third form of aluminum based on the pH and alkalinity of the raw water and the injection rate of the flocculant. Here, the concentration (abundance) of the first to third forms of aluminum can be determined from the respective ratios and the PAC injection rate.

  Next, the charge neutralization action and the crosslinking reaction action which are important in the aggregation reaction will be described. In charge neutralization, the turbid particles in the raw water are negatively charged, so they repel each other and exist stably in the water. That is, they account for the majority of what is reported as raw water turbidity. On the other hand, when a flocculant having a positive charge is added, the negative charge of the turbid particles is neutralized by the positive charge of the flocculant, so the repulsive force between the turbid particles decreases. At this time, micro flocs are formed by the flocculant and the turbid particles.

  When the injection rate of the flocculant is the most suitable amount for charge neutralization, the charged state of the micro floc is around ± 0 mV. In addition, when an appropriate amount or more of the flocculant is injected into the raw water, the charged state of the micro flocs tends to be positive. The form of aluminum that greatly contributes to charge neutralization is the second form of aluminum. The mixing pond model unit 201 calculates the injection amount of the second form of aluminum added from the ratio of the aluminum of the second form obtained based on the pH and alkalinity information of the raw water and the injection rate of the flocculant. . The mixing pond model unit 201 calculates a charge neutralization rate (%) which is a charge neutralization potential of the added flocculant based on the injection amount of the aluminum in the second form and the turbidity of the raw water.

  Since the sum of the turbid particles in the raw water can be estimated from the turbidity of the raw water, grasp the amount of the flocculant injected against the turbidity and the state of the second form of aluminum in the flocculant. Thus, the state of charge neutralization can be predicted. Here, an index representing the ratio of aluminum in the flocculant to the turbidity called ALT ratio (aluminum concentration in the injected flocculant relative to the raw water turbidity) is used. The aluminum in the flocculant used in the ALT ratio here may be the total aluminum concentration in the flocculant, but it is desirable to use the aluminum concentration limited to the aluminum in the second form. The mixing pond model unit 201 calculates the charge neutralization rate based on the aluminum concentration (ALT ratio) of the second form.

  The mixing pond model unit 201 obtains the amount of the first form aluminum and the amount of the third form aluminum from the ratio of the first form aluminum and the third form aluminum described above. The mixing pond model unit 201 calculates the surplus aluminum concentration at this point from the amount of the first form aluminum and the amount of the third form aluminum that did not contribute to the charge neutralization. The surplus aluminum concentration is a value including the third form of aluminum that acts on the crosslinking action. The mixing pond model unit 201 subtracts the aluminum concentration of the second form from the injection amount of the flocculant to obtain the surplus aluminum concentration.

  FIG. 4 is a diagram showing an outline of processing in the flock formation pond model unit 202 of the present embodiment. The flock formation pond model unit 202 models and uses the water quality reaction of the flock formation process in the flock formation pond 6. As shown in FIG. 4, the flock formation pond model unit 202 calculates the flock particle size, the flock density, and the residual aluminum concentration based on the surplus aluminum concentration and the raw water temperature.

  First, the floc formation pond model unit 202 arbitrarily determines the particle size after charge neutralization at the outlet of the mixing basin 5. For example, a value such as 50 μm is given as the initial value of the particle diameter. The flock formation pond model unit 202 sets a coefficient based on an index such as a charge neutralization rate or excess aluminum concentration, and calculates a particle diameter after charge neutralization. The floc formation pond model part 202 is a micro floc having a particle diameter after charge neutralization by the cross-linking action from the ratio of the surplus aluminum concentration to the concentration of the third form of aluminum having a great influence on the cross-linking action and the raw water turbidity. The degree of coarsening is calculated.

  The floc formation pond model unit 202 obtains the average particle size of the flocs coarsened in the flock formation pond 6. Since it is known that the floc growth rate depends on the water temperature, the flock formation pond model unit 202 uses the raw water temperature as an input signal to prevent the floc growth from proceeding when the water temperature is cooler than the average, When the water temperature is warmer than the average, the average particle diameter of the grown floc is obtained from the particle diameter after charge neutralization using a model in which the growth of floc proceeds. The floc formation pond model unit 202 calculates the floc density based on the surplus aluminum concentration and the surplus aluminum amount with respect to turbidity.

  The floc formation pond model unit 202 calculates the surplus aluminum concentration that has not finally reacted or was not used for processing as the residual aluminum concentration. The remaining aluminum may be a third form of aluminum, or a turbid component attached to the first form of aluminum or the second form of aluminum.

  FIG. 5 is a diagram showing an outline of processing in the sedimentation basin model unit 203 of the present embodiment. The sedimentation basin model unit 203 models and uses the water quality reaction of the sedimentation filtration process in the sedimentation basin 7. As shown in FIG. 5, the sedimentation basin model unit 203 calculates sedimentation turbidity (comparative treated water quality information or model treated water quality information) based on the floc particle size, the floc density, and the raw water turbidity. The sedimentation basin model unit 203 outputs the residual aluminum concentration as it is as the precipitation water aluminum concentration.

ｖｓ:沈降速度［ｍ／ｓ］又は［ｃｍ／ｓ］
Ｄ_ｐ：粒子径［ｍ］又は［ｃｍ］
ρ_ｐ：粒子の密度［ｋｇ／ｍ^３］又は［ｇ／ｃｍ^３］
ρ_ｆ：流体の密度［ｋｇ／ｍ^３］又は［ｇ／ｃｍ^３］
ｇ：重力加速度［ｍ／ｓ^２］又は［ｃｍ／ｓ^２］
η：流体の粘度[Ｐａ・ｓ]又は[ｇ／(ｃｍ・ｓ)] In the sedimentation basin 7, flocs coarsened by sedimentation are settled to become sludge. The sedimentation basin model part 203 calculates | requires the sedimentation speed of a floc based on the Stokes sedimentation formula as shown in the following (Formula 1) based on the particle size and density of a floc.

vs: sedimentation velocity [m / s] or [cm / s]
D _p : particle diameter [m] or [cm]
ρ _p : Particle density [kg / m ³ ] or [g / cm ³ ]
ρ _f : fluid density [kg / m ³ ] or [g / cm ³ ]
g: Gravity acceleration [m / s ² ] or [cm / s ² ]
η: fluid viscosity [Pa · s] or [g / (cm · s)]

  The sedimentation basin model unit 203 calculates the speed in the direction of flowing from the upstream to the downstream of the sedimentation basin 7 from the shape of the sedimentation basin 7 and the residence time based on the flow rate. The sedimentation basin model part 203 is based on the sedimentation velocity and the flow velocity of the sedimentation basin 7, and the floc that has flowed into the sedimentation basin 7 sinks to the bottom as sludge and overflows to reach the exit of the sedimentation basin 7 Fractionation. The sedimentation basin model unit 203 has an average particle size of flocs when flowing into the sedimentation basin 7 and has a distribution from small to large particle sizes, so that the smaller the particle size flocs, the more likely it will overflow. Model so as to increase the performance.

  For example, as shown in FIG. 5, in a graph 50 with the floc particle size as the horizontal axis and the floc amount as the vertical axis, the average particle size A of the floc is indicated by a broken line 51. In the graph 50, the sedimentation basin model unit 203 determines a distribution line 52 having a distribution according to the density of flocs around the average particle diameter A of flocs and an area according to the total floc amount. This utilizes the fact that the larger the floc density, the smaller the floc particle size variation, and the smaller the flock density, the larger the floc particle size variation. Such a tendency is considered to occur because the smaller the floc density, the easier it is for the flocs once formed to collapse, and the floc particle size tends to vary. Note that the dispersion of the particle size distribution may be obtained by incorporating the coefficient used when setting the particle size after charge neutralization.

  The sedimentation basin model unit 203 estimates the amount of floc having a particle size smaller than the threshold B according to the area of the region 54 surrounded by the broken line 53 and the distribution line 52. This threshold value B is divided into the floc that flows into the sedimentation basin 7 and sinks to the bottom as sludge and the one that overflows and reaches the exit of the sedimentation basin 7 from the sedimentation velocity and the flow velocity of the sedimentation basin 7. To do. The amount of floc estimated here indicates the amount of floc that overflows and reaches the outlet of the settling tank 7. The sedimentation basin model part 203 calculates | requires sedimentation water turbidity which is the turbidity of sedimentation water based on the quantity of the flock which reached the exit of the sedimentation basin 7. FIG. That is, the settling basin model unit 203 calculates the amount of flocs that have not grown sufficiently to overflow to the exit of the settling basin without settling as the sedimentation turbidity at the exit of the settling basin 7. The sedimentation basin model unit 203 defines that the residual aluminum concentration does not sink in the sedimentation basin 7, and calculates the value of the residual aluminum concentration as the sedimentation water aluminum concentration that is the aluminum concentration of the sedimentation water that exits from the outlet of the sedimentation basin 7. To do.

  The filter basin model unit 204 predicts a rise in the water level in the sand filter basin 8 based on the sedimentation turbidity and the sedimentation water aluminum concentration. In this prediction, the larger the sedimentation turbidity and the sedimentation aluminum concentration, the more easily clogging occurs in the sand filtration basin 8, and when clogging occurs, the water level of the sand filtration basin 8 tends to increase. We are using. The filter basin model unit 204 outputs information on the predicted filter pond water level rise to the parameter adjustment unit 24.

  The sludge information acquisition unit 205 obtains the total amount of flocculation and the flocculant based on the turbidity of the raw water and the injection rate of the flocculant based on the amount of floc that has settled, and uses this total amount as the amount of sludge generated in the sedimentation basin 7. calculate. The sludge information acquisition unit 205 calculates the sludge density by taking into account the density change due to the accumulation and consolidation, with the flock density at the time of the exit of the flock formation pond 6 as a reference. The sludge information acquisition unit 205 outputs to the parameter adjustment unit 24 sludge information including information on the amount of generated sludge and the density of the sludge.

  The parameter adjustment unit 24 may be configured to adjust the parameters in consideration of changes in the filtration pond water level rise and sludge information obtained from the water quality reaction model prediction unit 23 during parameter adjustment. In addition, the parameter adjustment unit 24 compares the information on the rise in the filtration pond water level and the sludge information obtained from the water quality reaction model prediction unit 23 with the actual measurement value stored in the data collection / storage unit 21 during parameter adjustment. The parameters may be changed based on the comparison result.

  The operation condition determination unit 206 determines whether or not the precipitation water turbidity and the precipitation water aluminum concentration received from the settling basin model unit 203 are predicted based on the injection rates of a plurality of patterns of chemicals within a predetermined range. To do. The operation condition determination unit 206 has the lowest chemical injection rate among the chemical injection rates corresponding to the precipitation water turbidity and precipitation water aluminum concentration received from the sedimentation basin model unit 203 within a predetermined range. Identify a model injection rate. The operation condition determination unit 206 outputs an operation condition including the model injection rate of the medicine identified as the lowest to the operation management support unit 25.

  The parameter adjustment unit 24 performs parameter setting and operation control for each processing unit in the water quality reaction model prediction unit 23. For example, the parameter adjustment unit 24 instructs the water quality reaction model prediction unit 23 to predict the precipitation water turbidity and the precipitation water aluminum concentration based on the injection rates of the chemicals in a plurality of patterns.

  When a new parameter is set in the water quality reaction model prediction unit 23, the parameter adjustment unit 24 stores information on the old parameter in the parameter history storage unit 240. The parameter adjustment unit 24 can refer to information on past parameters stored in the parameter history storage unit 240 at the time of parameter adjustment.

A specific example of a model in which the water quality reaction is more specifically expressed in the water quality reaction model prediction unit 23 will be described.
FIG. 6 is a diagram illustrating a specific example of a model obtained by more specifically formulating a water quality reaction in the water quality reaction model prediction unit 23 of the present embodiment. In FIG. 6, input information 61 is information input to the water quality reaction model prediction unit 23, and is information including water quality information of raw water and a chemical injection rate. The input information 61 includes a = raw water turbidity, b = raw water pH, c = raw water temperature, d = raw water alkalinity, e = flocculating agent injection rate, f = pH adjusting agent injection rate, and g = alkalinity adjusting agent injection rate. including.

  In addition to the information described above, the input information 61 may include information such as the type of flocculant and the basicity of aluminum in the case of an aluminum-based flocculant in order to use the water quality reaction in a mathematical model. . Further, the input information 61 may include a G value representing the agitation strength, in this case, the G value of the mixing basin 5 and the G value of the flock formation pond 6 as structural information of the plant. The input information 61 includes information necessary for a model obtained by formulating a water quality reaction.

The mixing pond / floc formation pond model 62 is a model used in the mixing pond model unit 201 and the flock formation pond model unit 202, and a function F _1A for obtaining the floc particle size A and a function F _1B for obtaining the floc density B. And a function F _1c for obtaining the residual aluminum concentration C. Function F _1A , Function F _1B and Function F _1c are: raw water turbidity a, raw water pH · b, raw water temperature c, raw water alkalinity d, flocculant injection rate e, pH adjuster injection rate f, and alkalinity adjuster injection This is a function having the input information 61 including the rate g as a variable. In the function F _1A , the function F _1B and the function F _1c , parameters such as coefficients for each variable are adjusted by the parameter adjustment unit 24 by simulation based on data stored in the data collection / storage unit 21.

The function F _1A for obtaining the floc particle size A calculates the charge neutralization rate and the surplus aluminum concentration based on the variables a, b, d, and e, and is charged based on the calculated charge neutralization rate and surplus aluminum concentration. This is a function that collectively performs a series of calculations of calculating the particle size of the floc after summation and calculating the floc particle size A based on the calculated particle size of the floc after charge neutralization and the variable c.

The function F _1B for obtaining the floc density B summarizes a series of calculations in which the surplus aluminum concentration is calculated based on the variables b, d, and e, and the floc density B is calculated based on the calculated surplus aluminum concentration and the variable a. It is a function to do.

The function F _{1C for obtaining} the residual aluminum concentration C is a series of calculations for calculating the surplus aluminum concentration based on the variables b, d, and e, and calculating the residual aluminum concentration C based on the calculated surplus aluminum concentration. The function to perform.

The settling basin model 63 includes a function F _2X for determining the settling water turbidity X that is a model used in the settling basin model unit 203, a function F _2Y for calculating the settling water aluminum concentration Y, and sludge that is a model used by the sludge information acquiring unit 205. And a function F _2Z for _obtaining the generation amount Z. The functions F _2X and F _2Y are functions having the flock particle diameter A, the flock density B, and the residual aluminum concentration C as variables. The function F _2Z is a function having the raw water turbidity a and the flocculant injection rate e as variables. In the function F _2X , the function F _2Y and the function F _2Z , parameters such as coefficients for each variable are adjusted by the parameter adjustment unit 24 through simulation based on data stored in the data collection / storage unit 21.

The function F _2X for determining the sediment water turbidity X is based on the variables A and B, and calculates the amount of floc that overflows and reaches the outlet of the sedimentation basin 7 and calculates the amount of floc that reaches the outlet of the sedimentation basin 7 Is a function that collectively performs a series of calculations of calculating the turbidity X of the precipitated water, which is the turbidity of the precipitated water. The function F _{2Y for obtaining} the precipitated water aluminum concentration Y is a function for calculating the precipitated water aluminum concentration Y based on the variable C. The function F _{2Z for obtaining} the sludge generation amount Z is a function for calculating the sludge generation amount Z based on the variables a and e.

The sand filtration pond model 64 has a function F ₃ for obtaining a rise in the water level of the sand filtration pond 8, which is a model used in the filtration pond model unit 204. The function F ₃ is a function having the precipitated water turbidity X and the precipitated water aluminum concentration Y as variables. In the function F ₃ , parameters such as coefficients for each variable are adjusted by the parameter adjustment unit 24 by simulation based on data stored in the data collection / storage unit 21. The function F ₃ for obtaining the water level rise in the sand filtration basin 8 is a function for predicting the water level rise in the sand filtration basin 8 based on the variables X and Y.

Next, the operation of the management support system 20 in this embodiment will be described.
FIG. 7 is a flowchart showing the operation of the management support system 20 in the present embodiment. As shown in FIG. 7, the data collection / storage unit 21 includes raw water water quality information, an injection rate of a chemical containing a flocculant, and raw water turbidity at the outlet of the sedimentation tank 7 measured by the sedimentation tank outlet turbidimeter 14. Data including the degree is acquired (step S71). The raw water quality information acquired by the data collection / storage unit 21 and the injection rate of the chemical containing the flocculant are input to the statistical calculation unit 22 and the water quality reaction model prediction unit 23.

  The statistical calculation unit 22 uses the water quality information of the raw water from the data collection / storage unit 21 and the injection rate of the coagulant to perform sedimentation water turbidity by statistical processing of the data stored in the data collection / storage unit 21. Statistically treated water quality information including is predicted (step S72). The statistical calculation unit 22 outputs the raw water quality information used for the prediction, the injection rate of the flocculant, and the prediction result to the parameter adjustment unit 24.

  The parameter adjustment unit 24 uses, for the water quality reaction model prediction unit 23, the input values (the raw water quality information and the flocculant injection rate) used when the statistical calculation unit 22 predicts the statistically treated water quality information. And instruct to predict comparatively treated water quality information including sedimentation turbidity. In response to an instruction from the parameter adjustment unit 24, the water quality reaction model prediction unit 23 uses the input value used when the statistical calculation unit 22 predicts the statistically treated water quality information, and compares the water quality reaction model prediction unit 23 with the precipitation water turbidity. The treated water quality information is predicted (step S73). The water quality reaction model prediction unit 23 outputs the processed water quality information for comparison, which is a prediction result, to the parameter adjustment unit 24.

  The parameter adjustment unit 24 compares the comparative treatment water quality information predicted by the water quality reaction model prediction unit 23 according to the instruction and the statistical treatment water quality information predicted by the statistical calculation unit 22, and the two values are different. It is determined whether or not (step S74). Here, when it is determined that the comparatively processed water quality information predicted by the water quality reaction model prediction unit 23 and the statistically processed water quality information predicted by the statistical calculation unit 22 are different (YES in step S74), parameters The adjustment unit 24 adjusts the parameters used in the water quality reaction model prediction unit 23 using the data of the data collection / storage unit 21, and sets new parameters in the water quality reaction model prediction unit 23 (step S75). When the parameter adjustment unit 24 finishes the process of step S75, the parameter adjustment unit 24 proceeds to the process of step S73.

  Here, when it is determined that the comparatively processed water quality information predicted by the water quality reaction model prediction unit 23 and the statistically processed water quality information predicted by the statistical calculation unit 22 are not deviated (NO in step S74), parameters The adjustment unit 24 instructs the water quality reaction model prediction unit 23 to predict model-processed water quality information including sedimentation turbidity when the flocculant injection rate is changed into a plurality of patterns. Thereby, the water quality reaction model prediction unit 23 predicts model treated water quality information including a plurality of precipitated water turbidities corresponding to a plurality of patterns of the flocculant injection rate. The operation condition determination unit 206 determines whether or not the model process water quality information predicted from the injection rates of the plurality of patterns of medicines and received from the settling basin model unit 203 is a value within a predetermined range. The model injection rate of the lowest coagulant among the injection rates of the coagulant corresponding to the model treated water quality information determined to be inside is specified. The operation condition determination unit 206 determines the operation condition including the specified model injection rate of the lowest drug and outputs the operation condition to the operation management support unit 25 (step S76).

  The operation management support unit 25 achieves the target water quality level based on the operation conditions from the statistical calculation unit 22 and the operation conditions from the water quality reaction model prediction unit 23 (operation conditions determined in step S76). The optimum operating conditions are generated within the range to be output and output to the plant operating unit 26 (step S77).

Thus, the management support system 20 can perform support for injecting a more appropriate amount when injecting the flocculant in the water purification plant 1. By saving the flocculant, the operation management of the water purification plant 1 can be reduced for the following three reasons.
(1) The amount of the flocculant used can be reduced.
(2) By reducing the possibility that the flocculant becomes excessive, the occurrence frequency of clogging in the sand filtration pond 8 can be reduced due to the influence of residual aluminum generated by the excessive flocculant. If the occurrence frequency of clogging is reduced, the frequency of washing the sand filtration basin 8 can be reduced.
(3) Sludge containing a large amount of residual aluminum due to the excess of the flocculant tends to have a high water content and requires a long time for drying treatment. However, by reducing the possibility that the flocculant becomes excessive, it is possible to prevent the sludge drying process from requiring a long time.

  The data collection / storage unit 21 in the management support system 20 stores the data measured by the flow meters 2a to 2d, the water temperature meter 10, the turbidity meter 11, the PH meter 12, the flow meter 13, and the sedimentation tank outlet turbidity meter 14. The configuration to be collected may be any configuration. For example, the data collection / storage unit 21 is configured to collect data measured by wired or wireless communication, and may be configured to communicate via a network such as the Internet, or configured to communicate using a dedicated line. It is done. For example, the data collection / storage unit 21 may be configured based on the Internet such as cloud computing. In this case, the management support system 20 installs a management support device including, for example, a statistical calculation unit 22, a water quality reaction model prediction unit 23, a parameter adjustment unit 24, and an operation management support unit 25, and the management support thereof. It is good also as a structure which utilizes the data collection / preservation | save part 21 on a cloud from an apparatus.

  In the above-described embodiment, each function unit in the management support system 20 is a software function unit, but may be a hardware function unit such as an LSI. Each functional unit in the management support system 20 may be configured using a PCL (programmable logic controller). In the embodiment described above, an example of a water purification plant process is shown as a process to be supported by the management support system 20, but any process can be applied as long as it is a process for purifying raw water using chemicals.

  According to at least one embodiment described above, it is possible to provide support for injecting a more appropriate amount of chemicals in a water treatment plant.

  When the functions in the management support system 20 described above are realized by software, a program for realizing those functions is recorded on a computer-readable recording medium, and the program is read into the computer system. May be executed. Here, the “computer system” includes hardware such as an OS (Operating System) and peripheral devices. The “computer-readable recording medium” refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD (Compact Disk) -ROM, or a storage device such as a hard disk built in the computer system. . Furthermore, “computer-readable recording medium” means a volatile memory (RAM) inside a computer system that becomes a server or client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. In addition, those holding programs for a certain period of time are also included.

In addition, the above program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line.
Further, the above program may be for realizing a part of the functions described above. Further, the program may be a so-called difference file (difference program) that can realize the above-described functions in combination with a program already recorded in the computer system.

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 1 ... Water purification plant, 3 ... Irrigation well, 4 ... Activated carbon contact pond, 5 ... Mixing pond, 6 ... Flock formation pond, 7 ... Sedimentation pond, 8 ... Sand filtration pond, 9 ... Purification pond, 20 ... Management support system, 21 Data collection and storage unit 22 Statistical calculation unit (statistical prediction unit) 23 Water quality reaction model prediction unit 24 Parameter adjustment unit 25 Operation management support unit 26 Plant operation unit 201 Mixing Pond model part, 202 ... Flock formation pond model part, 203 ... Sedimentation pond model part, 204 ... Filtration pond model part, 205 ... Sludge information acquisition part, 206 ... Operation condition determination part, 240 ... Parameter history storage part

Claims (7)

As the information acquired in the water treatment plant, water quality information on the quality of raw water, chemical information on injection of chemicals into the raw water, and treatment indicating the quality of treated water after treatment with the chemicals injected according to the chemical information While acquiring water quality information and statistically processing to identify the relationship between the chemical information and the treated water quality information, based on the identified relationship, the treated water quality within a predetermined range Statistics for obtaining a statistical injection rate that is a lower injection rate and predicting statistically processed water quality information based on the water quality information of the raw water to be predicted, the acquired statistical injection rate, and the specified relationship Predictive part,
Using a water quality reaction model that models the reaction caused by the injected chemicals, a model injection rate that provides a lower injection rate within a predetermined range of the water quality of the treated water is obtained, and the raw water to be predicted is obtained. A water quality reaction model prediction unit that predicts model-treated water quality information based on water quality information, the acquired model injection rate, and the water quality reaction model;
To control the chemical injection rate in the water treatment plant based on the statistical injection rate used by the statistical prediction unit for prediction and the model injection rate used by the water quality reaction model prediction unit for prediction. A management support unit for generating operation conditions for
Management support system comprising A parameter adjustment unit for changing parameters of the water quality reaction model in the water quality reaction model prediction unit;
In the parameter adjustment unit, the treated water quality information for comparison predicted by the water quality reaction model prediction unit based on the water quality information of the raw water to be predicted and the statistical injection rate and the statistically treated water quality information are different from each other. 2. The management support system according to claim 1, wherein the parameters of the water quality reaction model are changed. The water quality reaction model predicts the particle size and density of flocs generated by the injection of the chemical, and based on the particle size and density of the flocs, the turbidity of the raw water after the floc sedimentation treatment is determined in the model. It is a model to predict as treated water quality information,
The treated water quality information includes a measured value of the turbidity of the treated water after the floc sedimentation treatment,
The management support system according to claim 1, wherein the statistically treated water quality information is obtained by the statistical prediction unit predicting turbidity of treated water after the floc sedimentation process. When the comparative treated water quality information and the statistical treated water quality information are not deviated, the water quality reaction model prediction unit predicts a plurality of the model treated water quality information corresponding to the injection rate of the chemicals in a plurality of patterns. And
Among the injection rates of the chemicals in which the model process water quality information falls within a predetermined range, the minimum injection rate of the chemicals is specified as the model injection rate, and further includes a specifying unit for outputting.
The management support system according to claim 2, wherein the management support unit generates the operation condition based on the model injection rate output by the specifying unit.   The management support system according to any one of claims 1 to 4, wherein the water quality information includes turbidity of the raw water, water temperature of the raw water, pH of the raw water, and alkalinity of the raw water. As the information acquired in the water treatment plant, water quality information on the quality of raw water, chemical information on injection of chemicals into the raw water, and treatment indicating the quality of treated water after treatment with the chemicals injected according to the chemical information While acquiring water quality information and statistically processing to identify the relationship between the chemical information and the treated water quality information, based on the identified relationship, the treated water quality within a predetermined range Statistics for obtaining a statistical injection rate that is a lower injection rate and predicting statistically processed water quality information based on the water quality information of the raw water to be predicted, the acquired statistical injection rate, and the specified relationship Predictive steps,
Using a water quality reaction model that models the reaction caused by the injected chemicals, a model injection rate that provides a lower injection rate within a predetermined range of the water quality of the treated water is obtained, and the raw water to be predicted is obtained. A water quality reaction model prediction step for predicting model-treated water quality information based on the water quality information, the acquired model injection rate, and the water quality reaction model;
Based on the statistical injection rate used for prediction in the statistical prediction step and the model injection rate used for prediction in the water quality reaction model prediction step, for controlling the chemical injection rate in the water treatment plant A management support step for generating operating conditions;
A management support method. As the information acquired in the water treatment plant, water quality information on the quality of raw water, chemical information on injection of chemicals into the raw water, and treatment indicating the quality of treated water after treatment with the chemicals injected according to the chemical information While acquiring water quality information and statistically processing to identify the relationship between the chemical information and the treated water quality information, based on the identified relationship, the treated water quality within a predetermined range Statistics for obtaining a statistical injection rate that is a lower injection rate and predicting statistically processed water quality information based on the water quality information of the raw water to be predicted, the acquired statistical injection rate, and the specified relationship Predictive steps,
Using a water quality reaction model that models the reaction caused by the injected chemicals, a model injection rate that provides a lower injection rate within a predetermined range of the water quality of the treated water is obtained, and the raw water to be predicted is obtained. A water quality reaction model prediction step for predicting model-treated water quality information based on the water quality information, the acquired model injection rate, and the water quality reaction model;
To control the chemical injection rate in the water treatment plant based on the statistical injection rate used for prediction in the statistical prediction step and the model injection rate used for prediction in the water quality reaction model prediction step. A management support step for generating operating conditions for
Management support program to make a computer execute.

Images